The whole recording is offered first. The waveform shows loudness over time, and the shaded areas outside the handles are the parts being left out.

• Taller means louder, not higher. Pitch is not shown here at all.
• The marked gaps are silences of at least 0.7 seconds, found by comparing loudness in decibels rather than raw amplitude — which is why a quiet passage is not mistaken for a pause.
• If a recording opens with spoken introduction, the light language usually begins after one of those gaps. Drag the handles to start there.

What it does not mean. Nothing here detects which language is being spoken. The tool finds pauses, which are real and measurable. Where the boundary actually falls is your call, not its.

How this sample’s energy is divided across eight frequency bands, spaced logarithmically from 60 Hz to 4000 Hz.

• The eight bars are proportions of each other and add up to 100%. A tall bar means more of the sound’s energy sat in that frequency range.
• Anything below 60 Hz or above 4000 Hz is not counted, and the bands overlap slightly at their edges, so read these as a share across the eight bands rather than a share of everything in the recording.
• The pitch of a voice moves these numbers as much as what is being expressed does. This is why a sample is worth reading against your own baseline rather than against someone else’s.

What it does not mean. A zone is a frequency range. It is not an organ, a body part, or a place in the body. The chakra names label the ranges; they do not claim the sound acted on anything.

The same eight numbers, recomputed in short consecutive windows and stacked so the whole session reads left to right.

• Each vertical slice is one window. The window length scales with the recording, so a long sample does not become unreadably narrow.
• What is worth looking for is movement: a band that grows, fades, or trades places with another. One set of eight numbers averages a whole session flat, and this is where that flattening comes undone.
• A timeline that barely changes is a real result. It means the distribution held steady.

What it does not mean. Movement here is a change in the sound. It is not evidence of a change in a person.

The same analysis with every band edge doubled, which is exactly equivalent to playing the recording an octave lower.

• Content moves down the zone map together. Nothing is split and nothing is created.
• The places the two readings differ most are where content was sitting close to the edge of a band, so a shift of one octave sends it somewhere else.

What it does not mean. This adds no information the normal reading does not already contain. It is a second view of one measurement, not a second measurement.

A Lissajous figure — the curve traced by combining sine waves at the carrier frequencies of this sample’s strongest zones.

• The 2D figure uses your top two zones by energy, one driving each axis. The 3D figure uses the top three, one per axis.
• In the 3D figure each segment of the curve is colored for whichever zone’s axis it has travelled furthest along at that point, so you can see which of the trio is shaping each part of it.
• Drag either figure to rotate it. Two samples with similar dominant zones will trace similar figures, which is the point: it is a shape you can recognise.

What it does not mean. This is arithmetic driven by your measured frequencies. It is not a simulation of a physical response, and not a claim that sound traces these shapes in the body.

A valence score over words you type, using VADER — a published lexicon method that accounts for negation and intensifiers, so “I am not happy” scores negative.

• The compound score runs from −1 to +1. Around zero means the vocabulary was neutral, or that positive and negative words balanced out.
• Every word that carried valence is listed with its own score, so the number can be checked rather than taken on trust. If a word looks wrong to you, it probably is.
• This panel is for the comparison track — ordinary speech in any language — because a comparison needs something to compare against.

What it does not mean. It is a score over vocabulary. It does not measure the speaker’s emotional state, their health, or the effect of anything they heard.

What is kept, and what is not.

• The numbers are saved. The audio is not, unless you tick the box to keep it — that box is never ticked for you.
• Mark your first recording as a baseline: an ordinary, neutral use of your voice. Microphone, room and vocal pitch move the numbers as much as content does, so everything else is worth reading against it.
• The analysis version is stored with each sample. Samples measured by different engines are not silently compared.

What it does not mean. Saving a sample makes no claim about it. It records what was measured and when.

Everything you have logged, newest first.

• A sample kept with its audio can be played back here. One saved without audio keeps its numbers and loses nothing else.
• You can delete any sample, and you can delete your whole account yourself without asking anyone.

What it does not mean. Other participants cannot see your samples. As the person running the study, Angela can — that is stated plainly in the consent form rather than buried in it.

Your samples gathered together, so patterns across sessions can be looked at rather than remembered.

• Comparisons are most meaningful against your own baseline, on the same microphone, in the same room.
• Counts are counts. A larger number of samples makes a pattern easier to see; it does not make the pattern mean more.

What it does not mean. Nothing here is a result about a person. It is a summary of measurements.

The traditional names, colors and associations attached to the eight frequency ranges.

• This is presented as tradition and experiential language. It is here because it is the vocabulary practitioners already use, and because the zone colors have to come from somewhere.
• Everywhere else in this tool, a gold panel means the traditional framework and a blue one means a measurement. The line between them is drawn on every screen.

What it does not mean. No frequency, color or named source on this site has been demonstrated to produce a physical or medical effect. Nothing here diagnoses, treats or prevents anything.